Magnetron Ion Source Update, (Rosenstiel Fusor)

For the design and construction details of ion guns, necessary for more advanced designs and lower vacuums.
Post Reply
Jon Rosenstiel
Posts: 1367
Joined: Thu Jun 28, 2001 5:30 am
Real name: Jon Rosenstiel
Location: Southern California

Magnetron Ion Source Update, (Rosenstiel Fusor)

Post by Jon Rosenstiel » Mon Feb 19, 2007 4:13 am

Follow this link to my initial report: viewtopic.php?f=12&t=4989#p32817

Upgrades since my initial report:
1. Ceramtec feedthrough with 0.375” solid copper conductor.
2. NST full-wave rectified (with filter cap) ion source power supply.
3. Selection of NdFeB magnets to play with.

Photo #1. Ion source in place on fusor.
Photo #2 “Business end” of ion source.
Photo #3. Guts of ion source with N48 magnet and steel spacers.

I didn’t want to get into the complexities of water cooling, so I went for the next best thing, a feedthrough with a massive copper conductor that would (hopefully) conduct heat away from the ion source’s NdFeB magnet before it reached its Curie point. So far it’s worked well, but I haven’t really thrown any high power, extended runs at it yet.

I’ve played with a few different NdFeB magnet combinations. The original 0.375” square x 1” long N38SH magnet worked, but just barely. It would allow my fusor to run ~one mTorr lower pressure that normal, but would not operate in “beam injection mode”. (<7.5 mTorr, where glow discharge will not take place). Fusor operation was smooth with this magnet.

Second magnet on the menu was a 0.5” OD x 0.25” ID x 1” long N42 NdFeB magnet. The ion source would strike a plasma, but it didn’t produce enough ions to effect fusor operation.

Third on the menu were the “killer” 0.5” OD x 0.5” long N48 NdFeB magnets. The ion source actually worked too well when using two of these magnets. Just a few milliamps of ion source current was all that was needed to send my fusor into a uncontrollable tizzy. (It really wasn’t THAT uncontrollable, but one did need three hands to manage the pressure, power, and ion source current…all at the same time). Beam injection mode was no problem; I was getting neutrons (not a lot, but the He3 counter was chirping) at 40 kV, 0.6 mA at a pressure of (get this) 3.7 mTorr! No glow whatsoever, yet neutrons were being produced!

Using just one of the N48 magnets along with two steel spacers (one spacer at each end, as shown in photo #3) was a good compromise. Useable ion source current fell into the 8 to 20 mA range, and fusor control was a smoother. Still, fusor controllability is very dependant on ion source current. The higher the ion injection rate the more difficult it becomes to control the fusor.
The magnetron ion source does appear to increase fusor neutron output, (at the expense of controllability), but after running the following experiment, I’m not sure what is actually happening.


The setup:
HpGe detector aimed at fusor viewport (through 2.5mm Pb collimator) recording fusor Bremsstrahlung spectra. (See chart #4) Normally (ion source off) the point at which the Bremsstrahlung quickly falls off is very close to the fusor’s indicated grid voltage. In chart #4 that point (blue trace) is at 25 kV. The red trace (ion source on) is what I do not understand. The point at which the Bremsstrahlung falls off is almost 31 keV; about 7 kV above my fusor’s indicated grid voltage.

How the above results were obtained:

1. Initial fusor operating parameters: Viec = 25 kV, Iiec = 15 mA, pressure = 15.5 mTorr, TIER = 7.4E+04 n/s, variac @ 40 on dial. (Blue trace, chart #4)
2. Magnetron Ion Source powered on: Imag = 19 mA. Viec drops to 21 kV, Iiec rises to 25 mA, and pressure drops to 13.7 mTorr.
3. Fusor pressure is reduced via throttle valve to bring Iiec back down to 15 mA.
4. New, reduced pressure operating parameters: Viec = 24 kV, Iiec = 15 mA, pressure = 11.5 mTorr, Imag = 17 mA, Vmag = -1400 V, TIER = 1.9E+05 n/s, variac @ 40 on dial. (Red trace, chart #4)

By looking at the numbers the ion source is doing what it’s supposed to do. Chamber pressure is lower (15.5 to 11.5 mTorr), and neutron output is higher (7.4E+04 to 1.9E+05 n/s). But what’s with the Bremsstrahlung spectra? It’s telling me my fusor’s grid voltage has increased 6 kV…. while my instrumentation is telling me it has decreased 1 kV, and the variac setting hasn’t changed.

To add to the confusion, a capacitive pickup around my fusor’s hv cable indicates that the fusor is drawing considerably more power when the ion source is on. Ripple amplitude went from 9 kV (ion source off) to 33 kV (ion source on).

The only thing that I can think of that may be happening is that the ion source is somehow affecting the voltage / current phase relationship in my power supply’s transformer or ballast.

Thoughts, anyone?

Jon Rosenstiel
Attachments
Image2.gif
Image2.gif (11.69 KiB) Viewed 2429 times
Mag ion 007.jpg
Mag ion 007.jpg (60.82 KiB) Viewed 2429 times
Mag ion 001.jpg
Mag ion 001.jpg (50.4 KiB) Viewed 2429 times
Mag Ion on Fusor.jpg
Mag Ion on Fusor.jpg (62.79 KiB) Viewed 2429 times

tligon
Posts: 587
Joined: Tue Jan 02, 2007 6:58 pm
Real name:
Contact:

Re: Magnetron Ion Source Update, (Rosenstiel Fusor)

Post by tligon » Mon Feb 19, 2007 3:41 pm

Just that no glow does not mean no ions, as you have elegantly demonstrated. Ions emit light when they recombine with electrons ... in other words when they are lost. Ions at fusion energy levels in an ideal machine should not emit light at all.

What may be occurring is you've actually gotten the thing to run at a level where grid transparency is the dominant ion lifetime issue. In a normal glow discharge machine, ion lifetime is more likely limited by all the stray collisions and recombination going on.

Jon Rosenstiel
Posts: 1367
Joined: Thu Jun 28, 2001 5:30 am
Real name: Jon Rosenstiel
Location: Southern California

Re: Magnetron Ion Source Update, (Rosenstiel Fusor)

Post by Jon Rosenstiel » Mon Feb 19, 2007 4:14 pm

>Jon Rosenstiel wrote:
>But what’s with the Bremsstrahlung spectra? It’s telling me my >fusor’s grid voltage has increased 6 kV…. while my >instrumentation is telling me it has decreased 1 kV, and the >variac setting hasn’t changed.

My question "what's with the Bremsstrahlung spectra" should have been "what's with my instrumentation"? I have no doubt that the Bremsstrahlung spectra and the ripple waveform is telling the real story. My instrumentation is NOT showing me the true picture.

I think I need to ditch my 60 hZ, unfiltered power supply if I am to continue with these types of investigations. Gotta' get busy with that Spellman!

Jon Rosenstiel

User avatar
Richard Hull
Moderator
Posts: 11536
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Magnetron Ion Source Update, (Rosenstiel Fusor)

Post by Richard Hull » Mon Feb 19, 2007 5:48 pm

Jon,

You are probably correct in that now is the time for a stable regulated supply that is current limited.

Control issues abound in the fusor as we are traditionally operating it on the edge of glow mode. Attempting to augment the ion supply tends to screw with that already ragged edge. Joe Zambelli had the same issue with his machine and the heated ring filament with ionizer grid. Lots of control issues not found in the simple design.

We need a high gas load to keep the fuel level high to do simple fusion, otherwise, it is pump deep and precision control and pinpoint collide a sparser population of gas targets.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

denergyguy
Posts: 187
Joined: Tue Feb 13, 2007 1:36 am
Real name:

Re: Magnetron Ion Source Update, (Rosenstiel Fusor)

Post by denergyguy » Wed Feb 28, 2007 7:52 pm

Would a magetron from a microwave oven work for doing something like this?

Jon Rosenstiel
Posts: 1367
Joined: Thu Jun 28, 2001 5:30 am
Real name: Jon Rosenstiel
Location: Southern California

Re: Magnetron Ion Source Update, (Rosenstiel Fusor)

Post by Jon Rosenstiel » Tue May 15, 2007 4:50 am

I got to the bottom of the Bremsstrahlung spectra discrepancy mentioned in my original post. It was an artifact of my unfiltered, high ripple x-ray power supply.

I ran the same test using my newly constructed multiplier power supply and voila`, no difference.

Jon Rosenstiel
Attachments
Brem.gif
Brem.gif (16.27 KiB) Viewed 2429 times

Post Reply